Default-argument syntax ambiguity

[jariji]

Supposition.@check function commutative(a=intgen, b=intgen)
     add(a,b) == add(b,a)
end

Normally b=intgen would make intgen the default argument to the function, but here it's a generator for arguments to the function. What if I actually want to pass a generator as an argument? What does Supposition.@check function commutative(a=3, b=4) do -- does it error, requiring a Just(3) or what?

It seems ambiguous and potentially confusing. Perhaps it is worth considering alternative syntax, or maybe it'll grow on me.

[Seelengrab]

What if I actually want to pass a generator as an argument?

You mean if you want to use intgen itself inside of commutative? No need; just capture it, or pass a Data.Just(intgen).

What does Supposition.@check function commutative(a=3, b=4) do -- does it error, requiring a Just(3) or what?

That errors - as documented for @check, the arguments are expected to be Possibility objects:

help?> @check
  @check [key=val]... function...

[...]

  The arguments to the given function are expected to be generator strategies. The names
  they are bound to are the names the generated object will have in the test. These
  arguments will be shown should the property fail!

It seems ambiguous and potentially confusing. Perhaps it is worth considering alternative syntax, or maybe it'll grow on me.

I'm happy about any suggestions, but I'm struggling to come up with anything better than =, which clearly communicates assignment (of some form). :: doesn't work, since that's type assertion.

[jariji]

I'm struggling to come up with anything better than =, which clearly communicates assignment (of some form). :: doesn't work, since that's type assertion.

Yeah. ∈/in comes to mind but I'm not sure whether it's actually better.

[Seelengrab]

Right - those also have the disadvantage of being either unicode, or requiring whitespace surrounding the operator. That's not good either, because it means this is harder to type, and the whitespace makes the entire invocation quite a bit longer. Having to write out function for the inline-definition style is already quite bloaty, especially when a few config options are passed, so I'd rather avoid forcing the arguments to be passed in longer form as well.